Recent advances in the targeted delivery of radionuclides and radionuclide conjugation chemistry, and the increased availability of a-emitters appropriate for clinical use, have recently led to patient trials of radiopharmaceuticals labeled with a-particle emitters with very promising results. One of the stated goals (pillars) of the NIH is to develop more personalized medicine; in the realm of therapeutic nuclear medicine this translates as a need for more accurate personalized dosimetry. However, current dosimetry paradigms are poorly suited to a-particle therapy. This reality is reflected by the vast discrepancies between clinical (or experimental) toxicity and expected toxicity calculated using standard (absorbed fraction) organ-level modeling and dosimetry for (a) hematotoxicity in 223Ra therapy of bone metasteses and (b) renal toxicity seen in murine experiments in targeted a-particle immunotherapy. The objective of this work is to create a model more suited to a-particle emitters. After successful completion of the proposal, this model will provide explanations for experimental and clinical results not currently understood and also provide guidance for ongoing and future a-particle therapy of cancer. The range of the a-particles emitted by the radiopharmaceuticals is on the order of 50-80 microns. This scale is substantially smaller than: (a) the resolving power of clinical imaging detectors and modalities, and (b) the scale of human organs. This second is extremely important when one considers that the range of the emissions is actually often on the scale of the functional or anatomical sub-units of several key potentially dose-limiting organs at risk, including the kidney (functional sub-unit: th nephron), and the bone marrow (anatomical sub-unit of bone: the trabecula). The model proposed here will incorporate both sub-unit anatomical as well as dynamic modeling in order to accurately interpret the effects of a-particle therapy on potential dose-limiting organs for accurate dosimetry and treatment planning. As a first step simple geometrical models of the relevant sub-units (nephron, marrow cavity) will be created in GEANT4, a high-energy Monte Carlo software. The human anatomical information will be gathered from cadavers for anatomical accuracy and provide an array of parameters that reflect human diversity. The pharmacokinetic component will be developed in murine models and the conversion of macroscopically measured whole organ PK to specific sub-unit PK will be established. The translation to human assumes that the link between macroscopic and microscopic spatiotemporal relationship for a given agent measured in a pre- clinical model will apply to the human because the distribution of the agent to the different microscopic compartments should remain the same. Finally, the model will be tested in murine MTD experiments. Validation in the murine experiments combined with the high specificity regarding the potential for individual diversity in the human model will allow for accurate personalizable a-particle dosimetry in the clinic.